AI Red Teamer & Threat Researcher
Research prompt injection, jailbreak, and data-exfiltration techniques against LLM workflows.
Apply by emailThe work
Help us turn AI security into something teams can operate.
You will work closely with a small team across product, engineering, security, and customer delivery. We expect clear thinking, direct communication, and evidence that you can own useful work without layers of ceremony.
Send a short note, relevant work or links, your location, and why this role fits. We would rather read a specific, honest application than a polished generic one.